1. Field of the Invention
The present invention relates to an image processing apparatus, an image processing method, an image display apparatus using the same, and a mobile electronic device such as a mobile phone.
2. Description of the Related Art
Recently, technologies for displaying high definition images in the field of image display apparatus, for example, technologies for displaying precise CG (computer graphics) images and displaying other images more realistically have been developed. However, there are strong demands for displaying more highly gray-scaled and higher definition images by image display apparatuses than the images so far realized by the development of the image display technologies.
In image display apparatuses using digital signals as image signals, six to eight bits of an image signal are assigned to each of R (red), G (green) and B (blue) data. Due to the strong demands for displaying more highly gray-scaled and higher definition images, demands for using digital signals having a larger number of bits are expected to be increased in the future.
A main stream image display apparatus, which assigns six to eight bits of an image signal represented by 16 bits to each of R, G and B data, will be specifically described.
Image display data used in this type of image display apparatus can display 65536 colors since 216=65536. For displaying RGB color images with the 65536 color image display data, the 5-6-5 format is generally used. With this format, 5 bits are assigned to R data, 6 bits are assigned to G data, and 5 bits are assigned to B data. Thus, the image display data has 16 bits in total.
In a TFT liquid crystal display panel unit, 6 bits are assigned, as a value representing a gray scale level, to each of R, G and B data. Thus, the image display data has 18 bits in total.
An image signal corresponding to an input digital image signal is output for processing.
In order to match a 16-bit digital image signal with such a TFT liquid crystal display panel unit, gray scale compensation is performed for expanding R-pixel image display data and B-pixel image display data, which are each assigned 5 bits, to 6-bit image display data.
For such gray scale compensation, the following three systems are mainly used: (1) LSB (Least Significant Bit) fixing system, (2) MSB (most significant bit) repeating system, and (3) gray scale palette system.
According to (1) the LSB fixing system, one bit is added as the LSB to the 5-bit image display data so as to create 6-bit image display data. As the LSB, “1” or “0” is automatically set.
According to (2) the MSB repeating system, one bit is added as the LSB to the 5-bit image display data so as to create 6-bit image display data. As the LSB, the same value as the MSB is set, unlike (1) the LSB fixing system.
According to (3) the gray scale palette system, 5-bit image display data and 6-bit image display data is associated with each other by a palette referred to as a look-up table (LUT) or a conversion table. When one value represented by 5-bit image display data is input, the 6-bit image display data corresponding to that one value is output.
For improving the degree of gray scale, pseudo gray scale systems are available. Generally known pseudo gray scale systems include (4) dither system, (5) error diffusion system, and (6) FRC (Frame Rate Control) system.
According to (4) the dither system, reference pixel values are defined. In a certain area of pixels, discrete reference pixel values are used to display pixels having pixel values which are different from the reference pixel values. The ratio of the number of pixels having the non-reference pixel values with respect to the number of pixels having the reference pixel values (i.e., expression ratio) is first found. Then, in accordance with the expression ratio, the gradations between the reference pixel values (intermediate tones) are displayed.
According to (5) the error diffusion system, the pixel value of one pixel is quantized (or binarized). The difference between the quantized value and the original pixel value (i.e., quantization error) is assigned to the pixel values of surrounding pixels. Thus, the intermediate tones are displayed.
According to (6) the FRC system, reference values are defined. In a certain time period (e.g., one frame), discrete reference pixel values are used to display pixels having pixel values which are different from the reference pixel values. The ratio of a time period in which the pixels having the reference pixel values are displayed with respect to a time period in which the pixels having non-reference pixel values are displayed is found. Using this ratio, the gradations between the reference pixel values (intermediate tones) are displayed.
The systems (1), (2) and (3) have problems in terms of color reproducibility (gray scale reproducibility). This will be described hereinafter. In the following description, pixel value 00h of 5-bit image display data and 6-bit image display data is a pixel value corresponding to the darkest display. Pixel value 1Fh of 5-bit image display data and pixel value 3Fh of 6-bit image display data are each a pixel values corresponding to the brightest display.
(1) The LSB fixing system has the following problem. In the case where the gray scale expansion is performed by adding “0” to the LSB of color component image display data of the original image data, pixel value 1Fh (corresponding to the brightest display of 5-bit image display data) is converted into pixel value 3Eh. As a result, the brightest display (3Fh) which is possible with a display panel cannot be displayed. In the case where the gray scale expansion is performed by adding “1” to the LSB of color component image display data of the original image data, pixel value 00h (corresponding to the darkest display of 5-bit image display data) is converted into pixel value 01h of 6-bit image display data. As a result, the darkest display (00h) which is possible with a display panel cannot be displayed.
(2) The MSB repeating system has the following problem. Continuous pixel values of 5-bit image display data (for example, 0Fh and 10h) are converted into pixel values 1Eh and 21h of 6-bit image display data, which are not continuous. As a result,continuous brightness cannot be displayed.
(3) The gray scale palette system has the following problem. Once a palette for converting 5-bit image display data into 6-bit image display data is set, the same palette is used for all the images. The palette needs to be newly set each time different types of images (for example, graphic images, animation images and other general images) are to be displayed. This increases the workload of the user.
The problems of the systems (1), (2) and (3) are caused by the fact that the display capability (26=64 gradations) of the display panel of 6-bit image display data cannot be fully utilized. In the systems (1) and (2), either “0” or “1” is automatically added as the LSB. Therefore, the actual display is limited to 5-bit display (25=32 gradations). In the system (3), the number of types of data which can be included in a palette is 32.
When the display capability of the display panel cannot be fully utilized as above, the following problem may occur in addition to the above problems. In the systems (1), (2) and (3), the number of bits of image display signals is insufficient. Therefore, generally in an image, a portion, which should be represented by smooth graduation change, may undesirably be represented by stepwise stripes (such stripes will be referred to as “pseudo profile”).
The systems (4), (5) and (6) can improve the degree of gray scale capability of the image display apparatus when the number of gray scale bits of an input image signal is larger than the number of gray scale bits of the image display apparatus, i.e., when the gray scale capability of the image display apparatus is insufficient for the input image signal. However, when the number of gray scale bits of an input image signal is smaller than the number of gray scale bits of the image display apparatus, the display corresponding to the number of gray scale bits exceeding that of the input image signal is not performed, although the gray scale capability of the image display apparatus is sufficient for performing such a display. In other words, the gray scale capability of the image display apparatus cannot be fully utilized.